RefSeq Curation at NCBI - Adding Value to the Maize Genome Resources

NCBI created Zea mays RefSeq transcript-protein pairs and the associated NCBI Gene record from the INSDC maize accessions. This is a publicly available resource consisting of a non-redundant collection of sequence records for mRNA and the encoded protein; each pair with an associated NCBI Gene record. Curation of Zea mays RefSeq transcripts creates a more robust and accurate collection of genome resources for the maize research community.  Curation involves 1) identifying instances of multiple identical RefSeq transcripts for a particular gene and resolving by merging, 2) identifying instances of genes expressing transcript isoforms and creating the necessary RefSeq accessions, 3) identifying instances of RefSeq transcripts with sequence variations relative to the genome sequence and resolving by creating either a new version of the transcript accession or a new transcript accession, 4) identifying instances of structural and/or sequence deficiencies in the genome, 5) identifying instances where the annotation endeavors by NCBI and maizesequence.org differentially annotate a gene and 6) identifying instances of the genome sequence lacking a particular gene. Points four and six result in a record in the internal Genome Problem database at NCBI; the contents of which are shared with MaizeGDB. A consequence of points five and six is that NCBI may be able to still track the missing locus in the NCBI Gene resource and be able to provide a representative RefSeq transcript and protein record (i.e. one of the strong advantages to maintaining the transcript-based NM/NP dataset).